Proc. Natl. Acad. Sci. USAVol. 90, pp. 4017-4021, May 1993Biochemistry

A protein phosphatase related to the vaccinia virus VH1 is encodedin the genomes of several orthopoxviruses and a baculovirus

(phosphorylation/tyrosine/serine/raccoonpox virus)

DAVID J. HAKES*, KAREN J. MARTELL*, WEI-Guo ZHAOt, ROBERT F. MASSUNGt, JOSEPH J. ESPOSITOt,AND JACK E. DIXON*:*Department of Biological Chemistry and The Walther Cancer Institute, The University of Michigan, Ann Arbor, MI 48109-0606; and tCenters for DiseaseControl and Prevention, Poxvirus Section, 1600 Clifton Road, Mailstop G18, Atlanta, GA 30333

Communicated by Theodor 0. Diener, January 19, 1993 (received for review September 17, 1992)

ABSTRACT The vaccinia virus VH1 gene product is a dualspecificity protein phosphatase with activity against both phos-phoserine- and phosphotyrosine-containing substrates. We in-vestigated the potential presence of VH1 analogs in otherviruses. Hybridization and sequence data indicated that aphosphatase related to the VH1 phosphatase is highly con-served in the genomes of smallpox variola virus and otherorthopoxviruses. The open reading frames from the raccoon-pox virus and the smallpox variola virus Bangladesh majorstrain genomes encoding the VH1 analogs were sequenced andfound to be highly conserved with the vaccinia virus VH1. Anopen reading frame from the baculovirus Autographa califor-nica has sequence similarity to the VH1 phosphatase. The viralproteins appear to be structurally related to the cell cyclecontrol protein p8Ocdc2-. A recombinant phosphatase expressedfrom the baculovirus gene was found to share with the VH1phosphatase the ability to hydrolyze substrates that containedboth phosphoserine and phosphotyrosine.

Tyrosine phosphorylation is tightly regulated by the balanc-ing actions of protein-tyrosine kinases and protein-tyrosine-phosphatases (PTPases; protein-tyrosine-phosphate phos-phohydrolase, EC 3.1.3.48) (1). The PTPases catalyze phos-phate hydrolysis of phosphotyrosine and will not hydrolyzeeither phosphoserine- or phosphothreonine-containing sub-strates (2). Protein tyrosine phosphorylation plays a crucialrole in regulation of the eukaroytic cell cycle; alterations intyrosine phosphate levels can lead to a loss of control of cellgrowth (3, 4).

Bacteria are generally known to have very few tyrosine-phosphorylated proteins. Recently, however, a Yersinia spe-cies, which is the causative agent of the plague, was reportedto encode and express a PTPase (5). Deletional analysisindicated that the Yersinia PTPase was necessary for viru-lence (6), suggesting that the PTPase functions by dephos-phorylating tyrosine residues on host proteins critical formaintaining normal cellular function. Certain viruses alsoencode analogs of eukaryotic protein phosphatases (7). TheVH1 gene of vaccinia virus (HindIII-H, first open readingframe of strain WR) has been shown to encode a phosphatasethat has considerable amino acid identity with p8Ocdc25, aeukaryotic protein phosphatase important for regulation ofthe cell cycle (8), implying structural similarity between VH1and p8Ocdc25. Site-directed mutagenesis of the cysteine resi-due in the putative catalytic domain of either protein resultsin a total loss of detectable phosphatase activity (7). Themutant p8Ocdc25 also lacked biological function in an in vivoassay system (9). The most distinguishing feature of the171-amino acid VH1 phosphatase was that it hydrolyzed both

phosphoserine- and phosphotyrosine-containing proteins (7).Both substrates seem to be hydrolyzed by a common mech-anism since mutagenesis of a critical cysteine at the activesite of the phosphatase leads to a loss in hydrolytic activitytoward both substrates.Because Yersinia and vaccinia virus infections radically

affect the host cell cycle, we speculated that phosphatasesplay a role in pathogenesis (5, 7). To begin exploring thisconcept, we report here the identification and partial char-acterization of other phosphatase genes in orthopoxvirusesother than vaccinia virus, including smallpox variola virus,and in the baculovirus Autographa californica. We deter-mined analog VH1 sequences in the raccoonpox virus, whichis generally attenuated compared to vaccinia virus in small-pox vaccine relative virulence tests (10).§ Furthermore, all ofthe VH1-like phosphatase proteins we examined appear tohave conserved active sites based on our initial character-izations and the catalytic properties of this group of viralphosphatases.

MATERIALS AND METHODSMaterials. Sequenase was purchased from United States

Biochemical, casein was from Fluka, [y-32P]ATP (7000 Ci/mmol; 1 Ci = 37 GBq) was from New England Nuclear, srckinase was from Oncogene Sciences (Mineola, NY), and therandom primer labeling kit was from Boehringer Mannheim.The catalytic subunit of protein kinase A was a gift fromMichael Uhler (Department of Biological Chemistry, Uni-versity of Michigan). A. californica genomic DNA was a giftfrom Richard Jove (Department of Microbiology, Universityof Michigan).

Southern Blot Analysis. Genomic DNA from the followingorthopoxviruses was digested with HindIII, subjected toelectrophoresis, and bidirectionally transferred to nitrocel-lulose as described (11): variola virus (Bangladesh majorstrain and Garcia South American minor strain), vacciniastrain NYBH virus, monkeypox virus (Copenhagen strain),cowpox virus (Brighton strain), ectromelia virus, raccoonpoxvirus, volepox virus, and skunkpox virus (11, 12). The blotwas probed with a randomly primed 32P-labeled fragmentcontaining the coding sequence of the vaccinia VH1. Thefinal wash of the filter was highly stringent-i x standardsaline citrate at 65°C.

Sequencing of Raccoon VH1. The HindIII-E fragment (15kb) of raccoonpox virus genome DNA cloned into pUC18(13) hybridized with a probe containing the coding region ofthe vaccinia VH1 gene (Fig. 1). The fragment designated

Abbreviations: PlPase, protein-tyrosine-phosphatase; pNPP, p-ni-trophenyl phosphate; GST, glutathione S-transferase.ITo whom reprint requests should be addressed.§The sequence reported in this paper has been deposited in theGenBank data base (accession no. L13165).

4017

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 1,

202

1

Proc. Natl. Acad. Sci. USA 90 (1993)

MPV CPV ECT RCN VPX SKP

4. : ...

BkDa

69-

46-

30-

21.5-

14.3-

GST-VH1 VV RCN

FIG. 1. The VH1 gene is conserved in many orthopoxviruses. (A)

Genomes of certain members of the genus Orthopoxvirus were

digested with HindIII, subjected to electrophoresis on agarose gels,

transferred to nitrocellulose, and probed with the coding sequence of

VH1. Genomes are labeled as follows: VAR BSH (smallpox variola

virus Bangladesh major strain), VAR GAR (variola virus Garcia

South American minor strain), VV (vaccinia virus), MPV (monkey-

pox virus Copenhagen strain), CPV (cowpox virus Brighton strain),

ECT [ectromelia virus (mouse poxvirus)], RCN (raccoonpox virus),

VPX (volepox virus), and SKP (skunkpox virus). (B) Western blot

analysis of vaccinia virus VH1 (VV) and raccoonpox virus (RCN)

proteins with antibody prepared against the recombinant GST-VH1.

Molecular mass markers are as noted.

pUC18-RH15 was digested with BamHI. Digested fragments

were resolved by electrophoresis and then transferred to

nitrocellulose and probed with the VH1 coding sequence. An

'-2.8-kb hybridizing fragment was then subcloned into Blue-

script SK- (BS-RB2.8). BS-RB2.8 was digested with EcoRI

and probed as described above. An =l.3-kb hybridizing

fragment was subcloned into Bluescript SK- (BS-RE1.3).

BS-RE1.3 was digested with EcoRV and subjected to elec-

trophoresis. The band containing the vector and -950 bp of

the raccoonpox virus DNA was isolated and religated. The

fragment was sequenced by the Sanger dideoxynucleotide

chain-termination method (14).Sequence Analysis. DNA and protein sequence compari-

sons were performed with the FASTA (15, 16), PILEUP (17),

and BESTFIT (18) programs.

Protein Expression and Western Blot Analysis. PCR was

used to place BamHI and EcoRI sites on the ends of the

baculovirus VH1 coding region. Two synthetic oligonucleo-

tides (5'-TTTGGATCCATGTTTCCCGCGCGTTG-GCAC-3' and 5'-TTTGAATTCTTAAAATAAATCT-TGAACGTA-3') were used in the PCR with A. californica

viral genomic DNA as a template. The resulting fragment,

coding for ORF3, located on an EcoRI/Sal I fragment of the

HindIII-F region of the viral genome (19), was digested with

BamHI and EcoRI and subcloned into BamHI/EcoRI-digested pGEX-KT (20), producing the vector pGEX-KT-

BVH1. The recombinant vector was sequenced to ensure that

no sequence errors had been introduced by the PCR. The

glutathione S-transferase (GST; RX:glutathione R-transfer-

ase, EC 2.5.1.18) fusion protein was produced and purified as

described (21) with the exception that induction was per-formed at room temperature overnight to improve the solu-bility of the fusion protein.

Western blot analysis of vaccinia virus and raccoonpoxvirus proteins in infected cells was performed as described byAuperin et al. (22).Enzyme Activities. Phosphorylation of casein on serine and

tyrosine was carried out by using the catalytic subunit ofprotein kinase A and the recombinant src kinase, respec-tively, as described (7). Serine or tyrosine phosphorylatedcasein (1 nM; 100,000 cpm) was incubated with increasingamounts of GST-BVH1 in 50 mM imidazole, pH 7.0/0.1%2-mercaptoethanol at 37°C for 10 min. The reaction wasstopped by the addition of bovine serum albumin (80,ug) and50% trichloroacetic acid to a final concentration of 20%. Theprotein was precipitated on ice for 10 min. After centrifuga-tion, the amount of 32P released was determined by scintil-lation counting of the supernatant.

RESULTS AND DISCUSSIONIdentification of VH1 Hybridizing Sequences in Members of

the Genus Orthopoxvirus. A protein phosphatase (VH1), en-coded in the HindIII-H (leftward transcription) fragment ofthe vaccinia virus genome DNA, was shown to have activitytoward tyrosine and serine phosphate-containing substrates(7). Although the function of the phosphatase in viral repli-cation or pathogenesis is unknown, the structural similaritybetween VH1 and the cell cycle gene product p8ocdc25 hintsat the possibility that the viral gene may contribute to theobservation that vaccinia virus rapidly shuts down the cell(23). Efforts directed at inactivation of the VH1 gene havebeen unsuccessful (J.E.D. and S. Broyles, unpublished ob-servation). To gain additional insights into the importanceand function of VH1, we have examined other orthopoxvi-ruses as well as other viruses for VH1-like sequences.A Southern blot of HindIII-digested genomic DNA of

certain members of the genus Orthopoxvirus was probed athigh stringency with the coding sequence of VH1 (Fig. 1).Strong hybridization signals were detected under stringentconditions with the genomes of two strains of smallpoxvariola virus (i.e., Bangladesh major strain and the GarciaSouth American minor strain), monkeypox virus, cowpoxvirus, ectromelia virus, raccoonpox virus, skunkpox virus,and volepox virus. Intense signals were observed understringent hybridization conditions. In all of the orthopoxvi-ruses tested, the relative genomic location of the VH1 genecognate was conserved based on DNA maps (11, 13).

Recently, Knight et al. (13) showed the general genomicorganization of the orthopoxviruses is highly conserved;however, the HindlIl map of the raccoonpox virus is ratherdiverged from other orthopoxvirus DNA maps. The hybrid-ization signal in the HindIll digest of the raccoonpox virusgenomic DNA (Fig.1A, lane RCN) was less intense than thatof other orthopoxviruses tested. The possibility that thishybridization signal indeed corresponds to VH1-like DNAsequences was further analyzed with antibodies that recog-nize the vaccinia VH1 phosphatase. Infected cell lysateswere prepared for vaccinia virus and raccoonpox virus (11).Antibody against bacterially expressed VH1 protein wasshown to recognize proteins of -22 kDa in vaccinia virus-and raccoonpox virus-infected cells (Fig. 1B). The raccoon-pox virus VH1 analog (termed RVH1) was markedly lessreactive in Western blot analysis compared to the vacciniavirus VH1 protein. Addition of recombinant VH1 protein tothe hybridization solution used in Western blot analysisblocks the antibody-VH1 protein interaction (data notshown). In Fig. 1B, products of several sizes were seen withthe recombinant GST-VH1 protein. These products likelyrepresent proteolytic breakdown products of the fusion pro-

Asmallpox

IVAR VARBSH GAR VV

.mobdmil

mms

4018 Biochemistry: Hakes et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 1,

202

1

Proc. Natl. Acad. Sci. USA 90 (1993) 4019

tein. Collectively, both the Southern and Western blot anal-yses suggested to us that the RVH1 sequences were the leastconserved among the orthopoxviruses tested. We speculatedthat the degree of conservation of the RVH1 protein mightprovide insights into residues important for catalysis andfunction of the VH1-like phosphatases. Thus, to determinethe DNA sequence encoding the RVH1, a 15-kb genomicHindIII fragment from the raccoonpox virus was analyzed byrestriction enzyme digestion. An EcoRI/EcoRV fragmentthat hybridized to the vaccinia VH1 gene was subcloned andsequenced (see Materials and Methods for details).The DNA sequence encoding RVH1 was found to have

89% identity to the VH1 sequence (Fig. 2). At the proteinlevel, RVH1 showed 92% identity with the VH1 phosphatase.Only conservative amino acid substitutions were noted in theVH1 and RVH1 sequences. Although DNA maps suggestedthat vaccinia virus and raccoonpox virus genomes are themost divergent of the orthopoxviruses (13), the coding se-quences of VH1 and RVH1 are nearly identical. Residuesknown to be important for catalysis are invariant in the twoproteins (24). One can also speculate that since the raccoon-

pox virus phosphatase RVH1 appears to be the most diver-gent of the VH1-like sequences, then the VH1-like phos-phatases in the other orthopoxviruses are likely to have an

even higher degree of identity to VH1. This high degree ofconservation at the amino acid levels between VH1 andRVH1 and the likely possibility that VH1-like sequences invariola, cowpox, and other orthopoxviruses are even more

conserved suggest a vital role(s) for these phosphatases. Thesmallpox variola virus Bangladesh major strain VH1 analogwas sequenced and found to be 98% identical to VH1 at theamino acid level. All amino acids conserved between VH1and RVH1 are also conserved in the variola VH1, indicatingthat VH1 is not a single marker of virulence (J.J.E. andR.F.M., unpublished observation).

Identification of a VH1-Like Protein in Baculovirus. Toinvestigate whether the VH1-like phosphatases were found inother viruses, the coding region surrounding the active siteresidues ofVH1 and RVH1 was used to search the GenBankdata base. An open reading frame in baculovirus A. califor-nica (19) was identified that had significant sequence identityto the VH1 phosphatase. The entire coding sequences ofVH1, RVH1, and BVH1 were aligned (Fig. 2). Although thesizes of the three enzymes are very similar and the residuesnear the predicted active site cysteine are conserved, theBVH1 sequence shares only 20% identity with VH1 andRVH1. Taking into account conservative amino acidchanges, BVH1 shares 40% similarity to VH1 and RVH1.One can anticipate that ifBVH1 is indeed a phosphatase, thenthe invariant residues found in the three proteins are likely toplay critical roles in substrate binding, structure, or catalysis.

Table 1. Relative specific activities of VH1 and BVH1

Substrate VH1/BVH1

pNPP 40Casein (tyrosine) 55Casein (serine) >2500

Purification and Expression ofBVH1. The VH1 protein waspreviously shown to have activity toward tyrosine- and serine-phosphorylated substrates (7). The level of identity betweenBVH1 and VH1 raises the question of whether BVH1 wasindeed a phosphatase and underscored the importance ofdefining its substrate specificity. To address these issues, theBVH1 coding sequence was expressed in Escherichia coli asa fusion protein with GST. The fusion protein GST-BVH1 waspurified in a single step by glutathione agarose affinity chro-matography. The results of this purification are shown in Fig.3A. A crude lysate ofE. coli cells expressing GST-BVH1 (lane1) and purified fusion protein (lane 2) were analyzed bySDS/15% PAGE. The fusion protein was purified >95% asestimated from the single intensely staining protein bandshown in Fig. 3A (lane 2). The purified protein effectivelyhydrolyzed the widely used substrate, p-nitrophenyl phos-phate (pNPP) (Fig. 3B). The catalytic activity was inhibited byaddition of vanadate (Fig. 3B). Vanadate is a general PTPaseinhibitor and the vaccinia phosphatase is known to be sensitiveto this reagent (GST does not show any hydrolytic activitytoward pNPP). To examine the activity of BVH1 againsttyrosine- and serine-phosphorylated substrates, increasingamounts of BVH1 were incubated with equal concentrationsof32P-labeled serine- and tyrosine-phosphorylated casein. Fig.3C shows the 32P released from the tyrosine-phosphorylatedand the serine-phosphorylated casein. The BVH1 phosphataseefficiently dephosphorylated the tyrosine-phosphorylatedsubstrate but showed a low level of activity toward theserine-phosphorylated substrate. A comparison ofthe ratios ofspecific activities of BVH1 and VH1 is shown in Table 1.Using equal amounts of enzyme, VH1 has 40 times moreactivity against pNPP, 55 times the level of activity againsttyrosine-phosphorylated casein, and >2500 the level of activ-ity against serine-phosphorylated casein compared to BVH1.These results suggest that VH1 is more active than BVH1toward all three substrates examined. In addition, the ratio oftyrosine to serine phosphatase activity displayed by the twoproteins is radically different. VH1 shows a preference forphosphoserine-containing substrates while BVH1 hydrolyzesthe same substrates poorly and shows a preference for phos-photyrosine-containing casein as opposed to phosphoserine-containing casein. These apparent differences need to beinterpreted cautiously, however, since the substrates exam-ined here are only "model proteins" and may not be accurateindicators of the specific intracellular substrate(s) used by

Ih*DKKSLYKit LI?UDMEL 'fiMI TtN L ANEW (41)RV h 1 EDKKSLYKE L ST D IDHR P TMtU TNNV*L V YTK (41)

BVhl FPARWHN *QC QVIKDSNLICFK PL PELFA VTSEEDVWT (44)

V h 1 DXP 7EVtE1 V TTLP flI[RXPL EDTETR VI h u E APSSEVKF LI eKTSTT INII PM: |SBVh I EQIVKQNPSI GA IID NTSKYT GVHFLRAG'LLYKKIQVPGQ L

(82)(82)(90)

V h I FI K TID YtA s T3 AV A A (125)R'h1V DIUI TID ITA S JPL V 1A 2LA(I25)BVhl PPE IVQE I TVKE E PGM L TH I T Y \'CR (134)

V h I t Er.* (171V)R V h I 'UKES 1N FLTVTHSKK AT N I 1KIil71)BVhl HTLGIA QE AIDRFEKA HKI SN Q LF* (167)

FIG. 2. Alignment of VH1, RVH1, and BVH1. Amino acid sequences of the VH1-like phosphatases from vaccinia virus, raccoonpox virus,and baculovirus were aligned. Residues that are absolutely conserved in all three proteins are indicated by solid boxes. Those residues that areconserved between two of the three phosphatases are indicated by shaded boxes. Gaps were introduced by the program to maximize areas ofidentity.

Biochemistry: Hakes et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 1,

202

1

Proc. Natl. Acad. Sci. USA 90 (1993)

A

kDa106- '-80

1 2A Bvh 1

Cdc 14Vh 1

Rvh IMvh 1Yvh 1

Cd c 25

IG T nI N Y MI A K L G C LL A V N AMLIA V NW AMF Q I S

AT I

F AQ L S VTFFLVF EHSAHA PHL

B

32.5-

27.5

18.5

E3 0.70

0.60

0.50

° 0.40,C 0.30

0.20

0.10

C

r_

0.2Q0co00L

20

10

Casein (Serine)]ECassin (Tyrosine)I

0.50 r0.25

0.000 10 20 30

BVH1, ,mg

FIG. 3. (A) Purification of pGEX-KT-BVH1. The GST-BVH1fusion protein was expressed in E. coli and isolated by affinitychromatography using a glutathione agarose resin. Crude cell lysate(lane 1) and affinity-purified GST-BVH1 fusion protein (lane 2) weresubjected to electrophoresis on a SDS/15% polyacrylamide gel.Molecular mass markers were run as standards (sizes are indicated onthe left). (B) Hydrolysis ofp-nitrophenyl phosphate (pNPP) by BVH1.Hydrolysis of pNPP in the absence (m) and presence (A) of 10 ,uMsodium orthovanadate was monitored by increased absorbance at 410nm (y axis) with increasing amounts of protein (x axis). (C) Dephos-phorylation of phosphoserine- and phosphotyrosine-containingcasein. Release of 32p from serine-phosphorylated (m) and tyrosine-phosphorylated (o) casein is plotted (cpm x 10-3) (y axis) withincreasing amounts of BVH1 (x axis).

these enzymes. Comparisons of the hydrolytic rates and thesubstrate specificity need to be readdressed when the natu-rally occurring substrates for the two phosphatases areidentified. Nevertheless, it appears that the baculovirusphosphatase is similar to the vaccinia phosphatase in beingable to hydrolyze both phosphoserine- and phosphotyrosine-containing proteins. The dual specificity of these phos-phatases is in contrast to the specificity noted with all of the

BVhl

Cd:14

Vhl

RVh1

MVhl

YVh 1

CdC25

FIG. 4. (A) Active site signature sequences in the VH1 phospha-tase family. Amino acid sequences surrounding the active sites oftheVH1-like phosphatases from a baculovirus (BVH1), vaccinia virus(VH1), raccoonpox virus (RVH1), mice (MVH1), and yeast (YVH1)as well as the cell cycle regulatory proteins cdc14 and cdc25 fromSchizosaccharomyces pombe were aligned by using the PILEUPprogram. Residues that are conserved among four of the sevenproteins are indicated by solid boxes. (B) Family tree of VH1-likephosphatases. A representative family tree based on the degree ofamino acid identities was constructed by the PILEUP program.

mammalian and bacterial PTPases whose substrate specific-ities are restricted to the hydrolysis of tyrosine phosphate-containing protein.VH1-Like Proteins Comprise a Subfamily of Phosphatases.

When the VH1 and BVH1 phosphatase domains were used tosearch for related eukaroytic proteins, several were identi-fied. These include the cell cycle proteins cdc25 (8) and cdc14(25) in fission yeast; an open reading frame in budding yeastgenome adjacent to the DALI locus (26) known as YVH1[studies on YVH1 are presented elsewhere (27)]; and aserum-stimulatable immediate-early gene, 3CH134, in themouse (28) (termed MVH1, for mouse VH1-like protein, inthis paper). A comparison of the residues surrounding thepredicted active site cysteine residue in each enzyme isshown in Fig. 4A. All of the enzymes contain the conservedHis-Cys sequence as well as an arginine residue 6 amino acidsC-terminal to the cysteine. It is important to point out thatphosphatase activity has not been demonstrated for cdc14(25) or 3CH134 (28). With the exception of cdc14, all of thesephosphatases are more similar to one another than to otherPTPases at the active site. We therefore suggest that theseenzymes may constitute a subfamily ofphosphatases that arepresent in viruses, yeast, and mammals. The widespreadnature of these related phosphatases, their ability to alter orexploit the phosphorylation-dephosphorylation signal trans-duction pathway, and their potential importance as diseasecofactors underscore the importance ofunderstanding eventsassociated with their interactions in attenuated and patho-genic organisms.

We wish to thank Dr. Jackie Sheng for help in production andisolation of the VH1 antibodies. This work is supported by National

4020 Biochemistry: Hakes et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 1,

202

1

Proc. Natl. Acad. Sci. USA 90 (1993) 4021

Institute of Diabetes and Digestive and Kidney Diseases Grant18024. D.J.H. is supported by a fellowship from the AmericanCancer Society. K.J.M. is supported by a fellowship from theNational Alliance for Research on Schizophrenia and Depression.

1. Cool, D. E., Tonks, N. K., Charbonneau, H., Walsh, K. A.,Fischer, E. H. & Krebs, E. G. (1989) Proc. Natl. Acad. Sci.USA 86, 5257-5261.

2. Tonks, N. K., Diltz, C. D. & Fischer, E. H. (1988) J. Biol.Chem. 263, 6731-6737.

3. Hunter, T. & Cooper, J. A. (1985) Annu. Rev. Biochem. 54,897-930.

4. Fischer, E. H., Charbonneau, H. & Tonks, N. K. (1991) Sci-ence 253, 401-406.

5. Bliska, J. B., Guan, K. L., Dixon, J. E. & Falkow, S. (1991)Proc. Natl. Acad. Sci. USA 88, 1187-1191.

6. Bolin, I. & Wolf-Watz, H. (1988) Mol. Microbiol. 2, 237-245.7. Guan, K., Broyles, S. S. & Dixon, J. E. (1991) Nature (Lon-

don) 350, 359-362.8. Moreno, S. & Nurse, P. (1991) Nature (London) 351, 194.9. Gautier, J., Solomon, M. J., Booher, R. N., Bazan, J. F. &

Kirschner, M. W. (1991) Cell 67, 197-211.10. Esposito, J. J., Sumner, J. W., Brown, D. R., Ebert, J. W.,

Shaddock, J. H., He, B. X., Dobbins, J. G. & Fekadu, M.(1992) Vaccines 1992 (Cold Spring Harbor Lab., Plainview,NY), pp. 321-329.

11. Esposito, J. J. & Knight, J. C. (1985) Virology 143, 230-251.12. Cavallaro, K. F. & Esposito, J. J. (1992) Virology 190, 434-

439.

13. Knight, J. C., Goldsmith, C. S., Tamin, A., Regnery, R. L.,Regnery, D. C. & Esposito, J. J. (1992) Virology 190, 423-433.

14. Sanger, F., Nicklen, S. & Colson, A. R. (1977) Proc. Natl.Acad. Sci. USA 74, 5453-5467.

15. Higgins, D. G. & Sharp, P. M. (1989) Comput. Appl. Biosci. 5,151-153.

16. Pearson, W. R. & Lipman, D. J. (1988) Proc. Natl. Acad. Sci.USA 85, 2444-2448.

17. Feng, D. F. & Doolittle, R. F. (1987) J. Mol. Evol. 35, 351-360.18. Smith, T. F. & Waterman, M. S. (1981) Adv. Appl. Math. 2,

482-489.19. Tilakaratne, N., Hardin, S. E. & Weaver, R. F. (1991) J. Gen.

Virol. 72, 285-291.20. Hakes, D. J. & Dixon, J. E. (1992) Anal. Biochem. 202, 293-

298.21. Smith, D. B. & Johnson, K. S. (1988) Gene 67, 31-40.22. Auperin, D., Esposito, J., Lange, J., Bauer, S., Knight, J.,

Sasso, D. & McCormick, J. B. (1988) Virus Res. 9, 233-248.23. Joklik, W. K. (1968) Annu. Rev. Microbiol. 22, 359-390.24. Streuli, M., Krueger, N. X., Tsai, A. Y. M. & Saito, H. (1989)

Proc. Natl. Acad. Sci. USA 86, 8698-8702.25. Wan, J., Xu, H. & Grunstein, M. (1992) J. Biol. Chem. 267,

11274-11280.26. Buckholz, R. G. & Cooper, T. G. (1991) Yeast 7, 913-923.27. Guan, K., Hakes, D. J., Wang, Y., Park, H.-D., Cooper, T. G.

& Dixon, J. E. (1992) Proc. Natl. Acad. Sci. USA 89, 12175-12179.

28. Charles, C. H., Abler, A. S. & Lau, L. F. (1992) Oncogene 7,187-190.

Biochemistry: Hakes et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 1,

202

1